According to reports by Bloomberg’s Tony Capaccio and Defense News’ Sean Riley, the Defense Department’s Inspector General (IG) has concluded that the Pentagon’s efforts to deploy six massive enterprise resource planning (ERP) software programs are way over budget and years behind schedule. We are not talking about the specialized software programs to fly the Joint Strike Fighter or design a nuclear weapon. The ERP programs in question are all basically commercial, off-the-shelf, software programs designed to do routine back office functions such as accounting, billing and supply chain management. All large manufacturing and consumer sales companies such as General Motors, Proctor and Gamble, WalMart and McDonalds rely on such an ERP system. They have been in widespread use for more than twenty years.
The IG’s report says that the cost of fully implementing the six programs has more than doubled to $15.2 billion. The price of the Army’s Logistics Modernization Program (LMP), for example, has risen from $420 million to $3.9 billion; even then it will be twelve years behind schedule. The deployment date for the Air Force’s Defense Enterprise Accounting and Management System has slipped from 2009 to 2017 and the price has ballooned from nearly $420 million to around $2.2 billion. The new ERP systems are intended to replace a plethora of legacy programs improving DoD’s overall acquisition program and saving money. As it now stands it is hard to make the case that the increased costs will not negate most or all of the savings the six systems were expected to generate. Moreover, by the time these systems are fully deployed, all the way out to 2017, they are likely to be obsolete or require extensive and expensive upgrades.
What are the reasons for these debacles? The IG report quotes sources in one service as saying that there were “unanticipated complexities associated with implementing a commercial, off-the-shelf solution of this magnitude.” This is the excuse from the military that has repeatedly deployed halfway around the world to fight and win at least three major wars. The contractor tasked to implement the LMP system gave what is probably the best answer which is that it “delivers the services in accordance with government directed requirements.” Perhaps the contractor tried to tell the Army’s program managers that they were asking for the wrong things or imposing expensive and complicated requirements on a commercial system. Or, given the generally adversarial relationship between DoD’s acquisition corps and the private sector, the contractor just shook his or her head and kept quiet. The fact that the program manager that directed the contractor to take the LMP down the rabbit hole was gone when the problems emerged and his successor’s successor had to try and fix the mess must have also contributed to the problem. This is a pattern that is being repeated in the acquisition of major military platforms and weapons systems.
We could simply chalk this situation up to ignorant government program managers or greedy and conniving contractors. Or we could ask why it is that over and over again the government, in general, and DoD, in particular, screws up acquisitions that the private sector seems to do with ease? If past behavior is any indication, the IG’s report will result in senior Pentagon leadership ordering more investigations, increased oversight and lots of new reports. This would be precisely the wrong response. The problem is not too little oversight but too much, not too much communications between government buyer and commercial seller but too little.
Back in 1990, Stephen Kelman authored a brilliant analysis of the causes of problems such as cost overruns and schedule slippages in major government acquisition. He concluded, as a review of his book explained, “that the government makes bad decisions both because it is badly informed and because its procurement officials are encouraged to disregard some of the important information that they possess. The culprit in both instances is the system of full and open competition.”(*) Two case studies Kelman used to make his point were of attempts by government agencies — the Department of Agriculture and the Internal Revenue Service — to deploy massive computer systems. In both instances, the government acquired systems that while seeming to meet specifications did not function well, were extremely expensive to operate and had little growth potential.
Kelman’s fundamental conclusion is that no matter how tight the formulation of requirements, how detailed the contracts or intensive the oversight, the government cannot protect itself from the kind of disasters he and the DoD IG have documented because the problem is one of inexperience and lack of knowledge about complex systems, particularly involving rapidly changing technologies. But because the acquisition system is based on full and open competition, the government cannot partner with industry so as to define clearly and accurately what it wants to acquire and how the contract should be structured to achieve that goal. I would add that the government’s increasingly adversarial attitude towards the private sector and its drive for competition at all costs is only making a difficult situation worse.
In fairness to Kelman’s argument and to the government, not all acquisition officials are inexperienced and not all procurements go awry as have the six IT programs reviewed by the IG. But after more than forty years of attempting to reform the acquisition system and stop the cycle of cost increases and schedule slippages, perhaps it is time to consider a different approach.
(*) Jerry Mashaw, “The Fear of Discretion in Government Procurement,” Yale Journal on Regulation, January 1, 1991. A review of Steven Kelman, Procurement and Public Management: The Fear of Discretion and the Quality of Government Performance, Washington, D.C.: The AEI Press, 1990.
Find Archived Articles: